Among the greatest challenges of the 21st century is the reduction of disabilities that often evolve in the last decades of a long life and the adverse consequences associated with them. Severity of disability is often conventionally reported in population surveillance studies by counting the numbers of activity of daily living (ADL) and instrumental ADL (IADL) limitations which fail to express the types of activities that are limited. Thus, we propose to develop several approaches to staging activity limitation and test the predictive validity of each approach as alternative ways of aggregating ADL and IADL limitation for population surveillance. Staging is innovative because it conveys information about which activities are limited in addition to the overall severity of limitation. It therefore improves on he simple counts methodology which only captures overall severity. Baseline information will be obtained from new entrants into the 2005, 2006, 2007, and 2008 panels of the ongoing Medicare Current Beneficiary Survey (MCBS) (estimated N=16,000) Access to Care files. Adverse outcomes will be obtained by merging the Cost and Use files and will include inpatient hospitalization(s), skilled nursing facility (SNF), long-term care (LTC) facility use, and all-caus mortality with a follow-up time of 3 years. Aim 1 will establish alternative ways of defining stages in the baseline data that express both individuals' types of limitation and their severity based on difficulty or assistance needed. Aim 2 will determine which staging approaches best explain occurrences of each adverse outcome after adjusting for age, gender, and race. Aim 3 is theory driven and will explore how well ADL and IADL stages explain risks of adverse events in combination with separate domains consisting of sets of related variables describing individuals' social living circumstances and resources; health status; subjective physical and psychological well being; and functional expectations, respectively. Aim 4 is data driven and will culminate in the development and internal validation of parsimonious indices to predict each adverse outcome compiling the most predictive variables in each domain as identified in Aim 3. All analyses will account for the complex design of the MCBS. If successful, staging can be expected to serve as a tool for population surveillance and patient screening that can help stimulate future efforts to reduce the burdens of illness and disability in the US population by enabling disease and disability management and the tracking of economic, social, and health policies targeted to reduce adverse outcomes among people with ADL and IADL limitation. Knowledge of which ADLs and IADLs are limited (defined by stage thresholds) will have important implications for assessing quality of life and to the understanding of either the qualitative nature of difficulty experienced or ascertaining the types of assistance needed. This qualitative information is currently missing from ADL or IADL limitation counts.